A Prospective Test of the Dual-Pathway Model of Bulimic Pathology Mediating Effects of Dieting and Negative Affect Eric Stice Department of Psychology University of Texas at Austin ABSTRACT Because there have been few longitudinal investigations of integrative etiological theories of bulimia nervosa, this study prospectively tested the dual-pathway model using random regression growth curve models and data from a 3-wave community sample of adolescent girls ( N = 231). Initial pressure to be thin and thin-ideal internalization predicted subsequent growth in body dissatisfaction, initial body dissatisfaction predicted growth in dieting and negative affect, and initial dieting and negative affect predicted growth in bulimic symptoms. There was prospective evidence for most of the hypothesized mediational effects. Results are consistent with the assertion that pressure to be thin, thin-ideal internalization, body dissatisfaction, dieting, and negative affect are risk factors for bulimic pathology and provide support for the dual-pathway model. Approximately 4% of adolescent girls will suffer from bulimia nervosa ( Rand & Kuldau, 1992 ; Whitaker et al., 1990 ), making it one of the more prevalent psychiatric problems faced by this population. Bulimia nervosa is characterized by a persistent course, can result in serious medical complications, and is associated with comorbid psychopathology, including affective disorders, anxiety disorders, and substance abuse ( Fairburn, Cooper, Doll, Norman, & O'Connor, 2000 ; Garfinkel et al., 1995 ; Wilson, Heffernan, & Black, 1996 ). Additionally, bulimic pathology predicts future onset of obesity ( Stice, Cameron, Killen, Hayward, & Taylor, 1999 ), which results in elevated morbidity and mortality ( Dietz, 1998 ). Eating pathology also predicts onset of depression ( Stice, Hayward, Cameron, Killen, & Taylor, 2000 ), which foretells school dropout, interpersonal problems, unemployment, substance abuse, delinquency, and legal problems ( Gotlib, Lewsinsohn, & Seeley, 1998 ). An improved understanding of the factors that promote eating pathology would facilitate the design of more effective preventive and treatment interventions for this serious mental health problem. Although several longitudinal studies identified risk factors that predict bulimic pathology (e.g., Graber, Brooks-Gunn, Paikoff, & Warren, 1994 ; Killen et al., 1996 ; Leon, Fulkerson, Perry, & Early-Zald, 1995 ; Patton, Johnson-Sabine, Wood, Mann, & Wakeling, 1990 ), few tested integrative models of the mediational mechanisms by which these risk factors might work together to promote bulimic symptoms. Longitudinal studies that investigated mediational models have not been able to provide evidence of temporal precedence for each of the constituent pathways in the multivariate models. This latter omission probably resulted because there is no accepted test of mediation that generates prospective evidence for each of the criteria proposed by Baron and Kenny (1986) , specifically that (a) the independent variable predicts the mediator, (b) the mediator predicts the dependent variable, and (c) a significant relation between the independent and dependent variables becomes weaker when the effect of the mediator is controlled statistically. Accordingly, the aims of this study were to test a multivariate etiological model of bulimic pathology and to introduce a new approach for investigating mediation that can generate prospective evidence for each link in a mediational chain. Journal of Abnormal Psychology © 2001 by the American Psychological Association February 2001 Vol. 110, No. 1, 124-135 For personal use only--not for distribution. Page 1 of 2011/7/2001http://spider.apa.org/ftdocs/abn/2001/february/abn1101124.htmlThis study tested the dual-pathway model of bulimic pathology ( Stice & Agras, 1998 ), which represents a synthesis of the sociocultural ( Striegel-Moore, Silberstein, & Rodin, 1986 ), dietary ( Polivy & Herman, 1985 ), and affect regulation ( McCarthy, 1990 ) accounts of eating disorder. The dual-pathway model posits that internalization of the thin ideal espoused for women contributes to body dissatisfaction because this ideal is virtually unattainable (see Figure 1 ). It also hypothesizes that elevated pressure to be thin from family, peers, and the media fosters body dissatisfaction, because repeated messages that one is not thin enough likely promote discontent with one's body. Theoretically, this increased body dissatisfaction, in turn, fosters dieting and negative affect, which consequently increase the risk for bulimic pathology. 1 Body dissatisfaction is thought to lead to dieting because of the common belief that this is an effective weight control technique. Body dissatisfaction may also contribute to negative affect because appearance is a central evaluative dimension for women in our culture. Dieting, in turn, is thought to foster negative affect because of the failures that are often associated with weight control efforts and the impact of caloric deprivation on mood. Dieting is also theorized to result in a greater risk for bulimic pathology, because individuals may binge eat to counteract the effects of caloric deprivation. Further, dieting might promote binge eating because breaking strict dietary rules can result in disinhibited eating (the abstinence-violation effect). Negative affect might foster bulimic symptoms because it is commonly believed that eating provides comfort and distraction from negative emotions. Thus, this model posits that individuals may initiate bulimic behavior because of either extreme dieting or chronic negative affect or a combination of these factors. That is, either one of these two pathways may be sufficient to promote onset of bulimic pathology (both are not necessary). It might be noted that the dual-pathway model shares similarities with the spiral model ( Heatherton & Polivy, 1992 ), which focuses on body dissatisfaction, dieting, and negative affect, although the latter also asserts that low self-esteem promotes dieting and that there are reciprocal relations among dieting, negative affect, and bulimic behaviors. There is mounting evidence that pressure to be thin, thin-ideal internalization, body dissatisfaction, dieting, and negative affect are risk factors for bulimic pathology. Following Kraemer et al. (1997) , the term risk factor is used exclusively to refer to variables that have been found to predict
View Full Document
Unlocking...